Presentation 1999/1/21
Region Extraction by Competition of Multiple Active Contour Models
Yuki Matsuzawa, Susumu Horiguchi, Toru Abe,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Most of region extraction methods using active contour model deform the contour model according to the local information along contours. For this reason it is difficult to extract a region with complex shape or varying texture by these methods. In this article, to use the global information for region extraction, a method using competition of multiple active contours is proposed. In the proposed method, multiple active contour models are set in the object and the background regions respectively. For a region in each contour model, the likelihood is estimated from the distribution of feature value. By the likelihood-based competition between regions, an object region is extracted as a set of regions. The effectiveness of the proposed method was confirmed by experiments of artificial and real images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Region Extraction / Active Contour Model / Mixture Model / Fisher's Discriminant
Paper # PRMU98-183
Date of Issue

Conference Information
Committee PRMU
Conference Date 1999/1/21(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Region Extraction by Competition of Multiple Active Contour Models
Sub Title (in English)
Keyword(1) Region Extraction
Keyword(2) Active Contour Model
Keyword(3) Mixture Model
Keyword(4) Fisher's Discriminant
1st Author's Name Yuki Matsuzawa
1st Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology()
2nd Author's Name Susumu Horiguchi
2nd Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology
3rd Author's Name Toru Abe
3rd Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology
Date 1999/1/21
Paper # PRMU98-183
Volume (vol) vol.98
Number (no) 527
Page pp.pp.-
#Pages 8
Date of Issue